How To Judge If A Robocar Is Actually Good (Tesla Vs. Waymo)
Submitted by brad on Wed, 2025-02-05 10:36
Topic:
Tags:
In order to compare the two, it's important to figure out just how to judge the quality of a self-driving system.
Read more at Forbes.com in How To Judge If A Robocar Is Actually Good (Tesla Vs. Waymo)
Here is a video:
Comments
David Chase
Wed, 2025-02-05 17:58
Permalink
Possible ways to game (?) the safety numbers
Viewing this from the POV of a guy often on a bike, there's two things I do, that any of these companies could also do to improve their numbers. Whether these are fair or not, I'm not sure.
One, is to be strategic about routes. On a bike there a road I avoid (even though Our Fair City put a bike lane on it) because it's not actually that safe. It's multilane, meaning drivers must track more things, and also the multiple lanes tend to allow higher speeds. And, because it is lined with businesses, the parking is short-term, meaning that doors open comparatively often, relative to a street lined with apartments. A self-driving car company has the option of keeping track of intersections and routes with an unusually high crash rate, and adjusting routes to minimize that exposure, subject to travel time constraints. Human drivers don't do that.
The second, is the size of the vehicle. A smaller car can leave more clearance when passing other car or people, allowing more movement by the other party w/o a collision. A bicycle takes this to an extreme. For example, if I'm riding past a person on a sidewalk who gives the appearance of "perhaps I will step into the road", on a bicycle, I can leave 8 feet of clearance (in a 10-foot lane) without crossing the centerline. If I'm in a 7-foot wide car, I can only leave 3 feet. (Bike clearance samples: https://www.youtube.com/watch?v=hmq8kRcocFQ , https://www.youtube.com/watch?v=dLlyO3uqKWM )
Add new comment